A Small Temperature Probe with a Bright Future

Hank Hogan

To measure temperature on the nanoscale, you need a minuscule thermometer. Now a group of researchers has demonstrated that a fluorescent particle glued to the end of a sharp tip of an atomic force microscope might do the trick. By monitoring changes in the fluorescence of the particle, the team showed that the effect of a 200-nm-wide heater reached out 10 times its width.

Team member Lionel Aigouy, a researcher working for the Centre National de la Recherche Scientifique at École Supérieure de Physique et de Chimie Industrielles of Paris, noted that the fluorescent probe method was relatively easy to use and had a lateral resolution about equal to the size of the fluorescent particle. “It can work on any kind of device, and it can work in liquids,” he said.

Shown is an experimental setup for measuring temperature with nanoscale resolution. A fluorescent particle on the tip of an atomic force microscope operating in tapping mode is moved across a nickel nanoheater (inset). An infrared laser excites temperature-dependent two-photon fluorescence, whichis detected by a photomultiplier tube. Images reprinted with permission of Applied Physics Letters.

Other members of the team were from the University of Toulouse in France, from the University of Tokyo, and from École Nationale Supérieure de Chimie of Paris.

In measuring the heat produced by nanoscale heaters, the group opted not to use a resistive platinum/iridium wire for a thermal probe because such wires are too large. They also wanted a simpler fabrication technique than is currently possible with thermocouple sensors using tips made with carbon nanotubes. Therefore, they constructed a sensor from a fluoride glass particle codoped with ytterbium and erbium. After attaching this particle to the tip of a custom-made atomic force microscope, they illuminated the fluorescent particle with a 975-nm infrared laser from Coherent Inc. of Santa Clara, Calif.

Mapping the output of a nickel nanoheater can provide the topography of a 200-nm-wide strip, with height indicated by the scale (a), the thermally modulated fluorescence with alternating current flowing through the nanoheater (b) and room-temperature fluorescence (c).

Because of two-photon absorption, the particle emitted in the visible range. The researchers examined the 550-nm emission, which is sensitive to temperature, by placing a Hamamatsu photomultiplier tube behind a filter and mounting it, the filter and the required lenses on a microscope.

To test the probe, they constructed nanoheaters out of 40-nm-thick nickel stripes that were either 200 or 500 nm wide and 40 μm long. The heaters sat atop a 500-nm-thick layer of silicon dioxide, which in turn was positioned on top of a silicon wafer. The researchers sent alternating current through the nickel, measuring the fluorescence in lockstep with the current.

They moved the probe in tapping mode up to, across and then away from the heaters while they were on. They did the same with the heat off, producing a room temperature fluorescence reading for normalization.

The results showed that the nickel had the highest temperature and that the heat appeared up to 2 μm away because of thermal diffusion.

The group is working on improving the measurement technique by using better detectors, brighter and shorter-lived fluorescent particles, and operation in permanent contact mode. It also is going to extend the technique for use in liquids and for smaller devices.

“We plan to study the heating of smaller stripes, down to a few tens of nanometers in width,” Aigouy said.

Applied Physics Letters, Jan. 14, 2008, Vol. 92, 023101.